CVS: mhonarc/MHonArc/lib mhamain.pl,2.82,2.83 mhtxtplain.pl,2.46,2.47

Update of mhonarc/MHonArc/lib
Modified Files:
	mhamain.pl mhtxtplain.pl 
Log Message:
Bug #12314: Applied a fix that will work with versions of Perl with
the Encode module installed (i.e Perl >=5.8).  The fix converts
the input to perl's internal utf-8 so line breaking is done on characters
vs bytes.  The data is translated back after operation.

To avoid unnecessary translations, translation is not done on
us-ascii or iso-8859 sets, common 7/8-bit charsets.  Of course,
conversion could be avoided for other 8-bit sets, but charset check
may cause more overhead then translation, so only common ones are
checked for.


======================================================================
FILE: mhonarc/MHonArc/lib/mhamain.pl
<http://www.mhonarc.org/cgi-bin/viewcvs.cgi/*checkout*/mhonarc/MHonArc/lib/mhamain.pl?rev=2.83>

<http://www.mhonarc.org/cgi-bin/viewcvs.cgi/mhonarc/MHonArc/lib/mhamain.pl.diff?r1=2.82&r2=2.83&diff_format=h>
--- mhamain.pl	9 Jun 2005 01:12:59 -0000	2.82
+++ mhamain.pl	20 Jun 2005 04:25:18 -0000	2.83
@@ -30,5 +30,5 @@
 require 5;
 
-$VERSION = '2.6.12';
+$VERSION = '2.6.12+';
 $VINFO =<<EndOfInfo;
   MHonArc v$VERSION (Perl $] $^O)

======================================================================
FILE: mhonarc/MHonArc/lib/mhtxtplain.pl
<http://www.mhonarc.org/cgi-bin/viewcvs.cgi/*checkout*/mhonarc/MHonArc/lib/mhtxtplain.pl?rev=2.47>

<http://www.mhonarc.org/cgi-bin/viewcvs.cgi/mhonarc/MHonArc/lib/mhtxtplain.pl.diff?r1=2.46&r2=2.47&diff_format=h>
--- mhtxtplain.pl	27 May 2005 06:28:33 -0000	2.46
+++ mhtxtplain.pl	20 Jun 2005 04:25:18 -0000	2.47
@@ -318,5 +318,5 @@
     ## Check if max-width set
     if (($maxwidth > 0) && ($quote_style != Q_FLOWED)) {
-        $$data =~ s/^(.*)$/&break_line($1, $maxwidth)/gem;
+        break_lines($data, $charset, $maxwidth);
     }
 
@@ -438,8 +438,5 @@
                 if ($maxwidth > 0) {
                     # Fixed lines should be clipped to specified max.
-                    $line =~ s/\n\Z//;
-                    $line = break_line(
-                              $line, $maxwidth+
-                                  (length($line)-html_length($line))) . "\n";
+                    $line = break_lines($line, $charset, $maxwidth, 1);
                 }
                 if ($nonfixed) {
@@ -633,5 +630,40 @@
 ##---------------------------------------------------------------------------##
 
-sub break_line {
+sub break_lines {
+    my $data_in  = shift;
+    my $charset  = shift;
+    my $maxwidth = shift;
+    my $is_html  = shift; # hack for flowed processing
+    return unless $maxwidth > 0;
+
+    my $data = ref($data_in) ? $data_in : \$data_in;
+    my $do_encode = 0;
+    eval {
+        require Encode;
+        # Only translate to Perl's utf-8 if not an 8-bit charset.
+        # No harm if done for 8-bit, but try to avoid unnecesary
+        # overhead.  Translation done so line breaking is done
+        # on characters, not octets.
+        if ($charset !~ /us-ascii/i &&
+                $charset !~ /8859/i &&
+                !Encode::is_utf8($$data)) {
+            $$data = Encode::decode($charset, $$data);
+            $do_encode = 1;
+        }
+    };
+    $$data =~ s{
+        ^(.*)$
+    }{
+        _break_line($1, ($is_html ? $maxwidth+(length($1)-html_length($1))
+                                  : $maxwidth))
+    }gemx;
+    if ($do_encode) {
+        # Translate back to current encoding.
+        $$data = Encode::encode($charset, $$data);
+    }
+    $$data;
+}
+
+sub _break_line {
     my($str) = shift;
     my($width) = shift;


---------------------------------------------------------------------
To sign-off this list, send email to majordomo(_at_)mhonarc(_dot_)org with the
message text UNSUBSCRIBE MHONARC-COMMITS